DB2 FOR LUW / DB2FORLUW / 950 - DB2

When concurrent LOADs for the tables in the same tablespace are 
issued, one of them may fail with SQL3508N. This is a timing 
issue and rarely occurs. 
The typical scenario is as follows: 
- load A is starting up, and creates a directory 
  .../DB2mmmmm.PID 
- load B (for a table in the same tablespace mmmmm as load A) 
  is finishing, and as part of cleanup operation it deletes 
  the directory .../DB2mmmmm.PID 
- load A tries to create directory .../DB2mmmmm.PID/DB2nnnnn.OID 
  but the operation fails because directory .../DB2mmmmm.PID 
  now no longer exists 
- this causes load A to fail with SQL3508N 
 
At that time the db2diag.log shows following entries: 
 
2010-12-18-11.21.50.767852+540 E36268A903   LEVEL: Error(OS) 
PID     : 4968482           TID  : 25279    PROC : db2sysc 0 
INSTANCE: db2inst1          NODE : 000       DB  : SAMPLE 
APPHDL  : 0-702             APPID: *LOCAL.db2inst1.101218021239 
AUTHID  : DB2INST1 
EDUID   : 25279             EDUNAME: db2agent (SAMPLE) 0 
FUNCTION: DB2 UDB, oper system services, sqlomkdirp, probe:10 
MESSAGE : ZRC=0x860F000A=-2045837302=SQLO_FNEX "File not found." 
          DIA8411C A file "" could not be found. 
CALLED  : OS, -, mkdir 
OSERR  : ENOENT (2) "No such file or directory" 
DATA #1 : File name, 70 bytes 
/home/db2inst1/NODE0000/SQL00001/load/DB200006.PID/DB200007.OID 
DATA #2 : signed integer, 4 bytes 
488 
DATA #3 : signed integer, 4 bytes 
488 
DATA #4 : String, 105 bytes 
Search for ossError*Analysis probe point after this log entry 
for further self-diagnosis of this problem. 
 
2010-12-18-11.22.07.025892+540 I37172A562   LEVEL: Error 
PID     : 4968482           TID  : 25279    PROC : db2sysc 0 
INSTANCE: db2inst1          NODE : 000       DB  : SAMPLE 
APPHDL  : 0-702             APPID: *LOCAL.db2inst1.101218021239 
AUTHID  : DB2INST1 
EDUID   : 25279           EDUNAME: db2agent (SAMPLE) 0 
FUNCTION: DB2 UDB, database utilities, DIAG_ERROR, probe:0 
DATA #1 : String, 135 bytes 
LOADID: 25279.2010-12-18-11.20.55.178616.0 (6;7) 
open locfile for write failed! , -2029060079, 0, Detected in 
file:sqlulocf.C, Line:203 
 
2010-12-18-11.22.07.030718+540 I37735A601   LEVEL: Error 
PID     : 4968482           TID  : 25279    PROC : db2sysc 0 
INSTANCE: db2inst1          NODE : 000       DB  : SAMPLE 
APPHDL  : 0-702             APPID: *LOCAL.db2inst1.101218021239 
AUTHID  : DB2INST1 
EDUID   : 25279           EDUNAME: db2agent (SAMPLE) 0 
FUNCTION: DB2 UDB, database utilities, DIAG_ERROR, probe:0 
DATA #1 : String, 174 bytes 
LOADID: 25279.2010-12-18-11.20.55.178616.0 (6;7) 
/home/db2inst1/NODE0000/SQL00001/load/DB200006.PID/DB200007.OID/ 
load.loc, 
0, 0, Detected in file:sqlulocf.C, Line:204 
 
2010-12-18-11.22.07.040360+540 I38337A1080   LEVEL: Error 
PID     : 4968482            TID  : 25279    PROC : db2sysc 0 
INSTANCE: db2inst1           NODE : 000       DB  : SAMPLE 
APPHDL  : 0-702              APPID: *LOCAL.db2inst1.101218021239 
AUTHID  : DB2INST1 
EDUID   : 25279            EDUNAME: db2agent (SAMPLE) 0 
FUNCTION: DB2 UDB, database utilities, sqluWriteLocFile, 
probe:210 
MESSAGE : ZRC=0x8015006D=-2146107283=SQLU_CA_BUILT 
          "SQLCA has already been built" 
DATA #1 : LOADID, PD_TYPE_LOADID, 48 bytes 
LOADID: 25279.2010-12-18-11.20.55.178616.0 (6;7) 
DATA #2 : SQLCA, PD_DB2_TYPE_SQLCA, 136 bytes 
sqlcaid : SQLCA  sqlcabc: 136  sqlcode: -3508 sqlerrml: 70 
sqlerrmc: LOCATION FILE 1 
/home/db2inst1/NODE0000/SQL00001/load/DB200006. 
sqlerrp : SQLULOCF 
sqlerrd : (1) 0x8015006D   (2) 0x00000000   (3) 0x00000000 
          (4) 0x00000000   (5) 0x00000000   (6) 0x00000000 
sqlwarn : (1)      (2)      (3)      (4)        (5)      (6) 
          (7)      (8)      (9)      (10)        (11) 
sqlstate:

**************************************************************** 
* USERS AFFECTED:                                              * 
* ALL                                                          * 
**************************************************************** 
* PROBLEM DESCRIPTION:                                         * 
* To avoid timing issues with concurrent loads a retry count   * 
* will be introduced so if we fail on accessing the            * 
* directories for Load control files we try again before       * 
* failing the load.                                            * 
**************************************************************** 
* RECOMMENDATION:                                              * 
* Upgrade to DB2 Version 9.5 and Fix Pack 8                    * 
****************************************************************

Issue LOADs for the same tablespace serially.

Problem was first fixed in DB2 Version 9.5 and Fix Pack 8

not known / see Local fix

Date  - problem reported    :
Date  - problem closed      :
Date  - last modified       :

07.01.2011
27.06.2011
27.06.2011

9.5.